Cosmos DB
Azure Cosmos DB is Microsoft’s proprietary globally-distributed, multi-model database service ‘for managing data at planet-scale’ launched in May 2017. It is schema-agnostic, horizontally scalable and generally classified as a NoSQL database. Internally, Cosmos DB stores ‘items’ in ‘containers’, with these 2 concepts being surfaced differently depending on the API used (these would be ‘documents’ in ‘collections’ when using the MongoDB-compatible API, for example). Containers are grouped in ‘databases’, which are analogous to namespaces above containers. Containers are schema-agnostic, which means that no schema is enforced when adding items. …

Complex Systems
Complex systems present problems both in mathematical modelling and philosophical foundations. The study of complex systems represents a new approach to science that investigates how relationships between parts give rise to the collective behaviors of a system and how the system interacts and forms relationships with its environment. The equations from which models of complex systems are developed generally derive from statistical physics, information theory and non-linear dynamics and represent organized but unpredictable behaviors of natural systems that are considered fundamentally complex. The physical manifestations of such systems are difficult to define, so a common choice is to identify ‘the system’ with the mathematical information model rather than referring to the undefined physical subject the model represents. Such systems are used to model processes in computer science, biology, economics, physics, chemistry, and many other fields. It is also called complex systems theory, complexity science, study of complex systems, sciences of complexity, non-equilibrium physics, and historical physics. A variety of abstract theoretical complex systems is studied as a field of mathematics. The key problems of complex systems are difficulties with their formal modelling and simulation. From such a perspective, in different research contexts complex systems are defined on the basis of their different attributes. Since all complex systems have many interconnected components, the science of networks and network theory are important aspects of the study of complex systems. A consensus regarding a single universal definition of complex system does not yet exist. For systems that are less usefully represented with equations various other kinds of narratives and methods for identifying, exploring, designing and interacting with complex systems are used. …

Diagonalwise Refactorization
Depthwise convolutions provide significant performance benefits owing to the reduction in both parameters and mult-adds. However, training depthwise convolution layers with GPUs is slow in current deep learning frameworks because their implementations cannot fully utilize the GPU capacity. To address this problem, in this paper we present an efficient method (called diagonalwise refactorization) for accelerating the training of depthwise convolution layers. Our key idea is to rearrange the weight vectors of a depthwise convolution into a large diagonal weight matrix so as to convert the depthwise convolution into one single standard convolution, which is well supported by the cuDNN library that is highly-optimized for GPU computations. We have implemented our training method in five popular deep learning frameworks. Evaluation results show that our proposed method gains $15.4\times$ training speedup on Darknet, $8.4\times$ on Caffe, $5.4\times$ on PyTorch, $3.5\times$ on MXNet, and $1.4\times$ on TensorFlow, compared to their original implementations of depthwise convolutions. …

Neural Nearest Neighbors Block (N3 Block)
Non-local methods exploiting the self-similarity of natural signals have been well studied, for example in image analysis and restoration. Existing approaches, however, rely on k-nearest neighbors (KNN) matching in a fixed feature space. The main hurdle in optimizing this feature space w.r.t. application performance is the non-differentiability of the KNN selection rule. To overcome this, we propose a continuous deterministic relaxation of KNN selection that maintains differentiability w.r.t. pairwise distances, but retains the original KNN as the limit of a temperature parameter approaching zero. To exploit our relaxation, we propose the neural nearest neighbors block (N3 block), a novel non-local processing layer that leverages the principle of self-similarity and can be used as building block in modern neural network architectures. We show its effectiveness for the set reasoning task of correspondence classification as well as for image restoration, including image denoising and single image super-resolution, where we outperform strong convolutional neural network (CNN) baselines and recent non-local models that rely on KNN selection in hand-chosen features spaces. …

Ultra-Scalable Spectral Clustering (U-SPEC)
This paper focuses on scalability and robustness of spectral clustering for extremely large-scale datasets with limited resources. Two novel algorithms are proposed, namely, ultra-scalable spectral clustering (U-SPEC) and ultra-scalable ensemble clustering (U-SENC). In U-SPEC, a hybrid representative selection strategy and a fast approximation method for K-nearest representatives are proposed for the construction of a sparse affinity sub-matrix. By interpreting the sparse sub-matrix as a bipartite graph, the transfer cut is then utilized to efficiently partition the graph and obtain the clustering result. In U-SENC, multiple U-SPEC clusterers are further integrated into an ensemble clustering framework to enhance the robustness of U-SPEC while maintaining high efficiency. Based on the ensemble generation via multiple U-SEPC’s, a new bipartite graph is constructed between objects and base clusters and then efficiently partitioned to achieve the consensus clustering result. It is noteworthy that both U-SPEC and U-SENC have nearly linear time and space complexity, and are capable of robustly and efficiently partitioning ten-million-level nonlinearly-separable datasets on a PC with 64GB memory. Experiments on various large-scale datasets have demonstrated the scalability and robustness of our algorithms. The MATLAB code and experimental data are available at https://…/330760669. …

ADAGE
The ability to generalize across visual domains is crucial for the robustness of visual recognition systems in the wild. Several works have been dedicated to close the gap between a single labeled source domain and a target domain with transductive access to its data. In this paper we focus on the wider domain generalization task involving multiple sources and seamlessly extending to unsupervised domain adaptation when unlabeled target samples are available at training time. We propose a hybrid architecture that we name ADAGE: it gracefully maps different source data towards an agnostic visual domain through pixel-adaptation based on a novel incremental architecture, and closes the remaining domain gap through feature adaptation. Both the adaptive processes are guided by adversarial learning. Extensive experiments show remarkable improvements compared to the state of the art. …

IL-Net
Deep neural networks (DNN) excel at extracting patterns. Through representation learning and automated feature engineering on large datasets, such models have been highly successful in computer vision and natural language applications. Designing optimal network architectures from a principled or rational approach however has been less than successful, with the best successful approaches utilizing an additional machine learning algorithm to tune the network hyperparameters. However, in many technical fields, there exist established domain knowledge and understanding about the subject matter. In this work, we develop a novel furcated neural network architecture that utilizes domain knowledge as high-level design principles of the network. We demonstrate proof-of-concept by developing IL-Net, a furcated network for predicting the properties of ionic liquids, which is a class of complex multi-chemicals entities. Compared to existing state-of-the-art approaches, we show that furcated networks can improve model accuracy by approximately 20-35%, without using additional labeled data. Lastly, we distill two key design principles for furcated networks that can be adapted to other domains. …

Adversarial Machine Learning (AML)
Adversarial machine learning is the formal name for studying what happens when conceding even a slightly more realistic alternative to assumptions of these types (harmlessly called ‘relaxing assumptions’ ….
Adversarial Machine Learning …

Spatial
A high-level language for programming accelerators (FPGA). …

AutoRL
Many continuous control tasks have easily formulated objectives, yet using them directly as a reward in reinforcement learning (RL) leads to suboptimal policies. Therefore, many classical control tasks guide RL training using complex rewards, which require tedious hand-tuning. We automate the reward search with AutoRL, an evolutionary layer over standard RL that treats reward tuning as hyperparameter optimization and trains a population of RL agents to find a reward that maximizes the task objective. AutoRL, evaluated on four Mujoco continuous control tasks over two RL algorithms, shows improvements over baselines, with the the biggest uplift for more complex tasks. The video can be found at: \url{https://youtu.be/svdaOFfQyC8}. …

Dynamic Planning Network (DPN)
We introduce Dynamic Planning Networks (DPN), a novel architecture for deep reinforcement learning, that combines model-based and model-free aspects for online planning. Our architecture learns to dynamically construct plans using a learned state-transition model by selecting and traversing between simulated states and actions to maximize valuable information before acting. In contrast to model-free methods, model-based planning lets the agent efficiently test action hypotheses without performing costly trial-and-error in the environment. DPN learns to efficiently form plans by expanding a single action-conditional state transition at a time instead of exhaustively evaluating each action, reducing the required number of state-transitions during planning by up to 96%. We observe various emergent planning patterns used to solve environments, including classical search methods such as breadth-first and depth-first search. Learning To Plan shows improved data efficiency, performance, and generalization to new and unseen domains in comparison to several baselines. …

VERtex Similarity Embedding (VERSE)
Embedding a web-scale information network into a low-dimensional vector space facilitates tasks such as link prediction, classification, and visualization. Past research has addressed the problem of extracting such embeddings by adopting methods from words to graphs, without defining a clearly comprehensible graph-related objective. Yet, as we show, the objectives used in past works implicitly utilize similarity measures among graph nodes. In this paper, we carry the similarity orientation of previous works to its logical conclusion; we propose VERtex Similarity Embeddings (VERSE), a simple, versatile, and memory-efficient method that derives graph embeddings explicitly calibrated to preserve the distributions of a selected vertex-to-vertex similarity measure. VERSE learns such embeddings by training a single-layer neural network. While its default, scalable version does so via sampling similarity information, we also develop a variant using the full information per vertex. Our experimental study on standard benchmarks and real-world datasets demonstrates that VERSE, instantiated with diverse similarity measures, outperforms state-of-the-art methods in terms of precision and recall in major data mining tasks and supersedes them in time and space efficiency, while the scalable sampling-based variant achieves equally good results as the non-scalable full variant. …

Semi-Supervised Deep Kernel Learning (SSDKL)
Large amounts of labeled data are typically required to train deep learning models. For many real-world problems, however, acquiring additional data can be expensive or even impossible. We present semi-supervised deep kernel learning (SSDKL), a semi-supervised regression model based on minimizing predictive variance in the posterior regularization framework. SSDKL combines the hierarchical representation learning of neural networks with the probabilistic modeling capabilities of Gaussian processes. By leveraging unlabeled data, we show improvements on a diverse set of real-world regression tasks over supervised deep kernel learning and semi-supervised methods such as VAT and mean teacher adapted for regression. …

Self-Attention Generative Adversarial Network (SAGAN)
In this paper, we propose the Self-Attention Generative Adversarial Network (SAGAN) which allows attention-driven, long-range dependency modeling for image generation tasks. Traditional convolutional GANs generate high-resolution details as a function of only spatially local points in lower-resolution feature maps. In SAGAN, details can be generated using cues from all feature locations. Moreover, the discriminator can check that highly detailed features in distant portions of the image are consistent with each other. Furthermore, recent work has shown that generator conditioning affects GAN performance. Leveraging this insight, we apply spectral normalization to the GAN generator and find that this improves training dynamics. The proposed SAGAN achieves the state-of-the-art results, boosting the best published Inception score from 36.8 to 52.52 and reducing Frechet Inception distance from 27.62 to 18.65 on the challenging ImageNet dataset. Visualization of the attention layers shows that the generator leverages neighborhoods that correspond to object shapes rather than local regions of fixed shape. …

Mutation-Selection Equilibrium
We propose a class of evolutionary models that involves an arbitrary exchangeable process as the breeding process and different selection schemes. In those models, a new genome is born according to the breeding process, and then a genome is removed according to the selection scheme that involves fitness. Thus the population size remains constant. The process evolves according to a Markov chain, and, unlike in many other existing models, the stationary distribution — so called mutation-selection equilibrium — can be easily found and studied. The behaviour of the stationary distribution when the population size increases is our main object of interest. Several phase-transition theorems are proved. …

Sinkhorn AutoEncoder (SAE)
Optimal Transport offers an alternative to maximum likelihood for learning generative autoencoding models. We show how this principle dictates the minimization of the Wasserstein distance between the encoder aggregated posterior and the prior, plus a reconstruction error. We prove that in the non-parametric limit the autoencoder generates the data distribution if and only if the two distributions match exactly, and that the optimum can be obtained by deterministic autoencoders. We then introduce the Sinkhorn AutoEncoder (SAE), which casts the problem into Optimal Transport on the latent space. The resulting Wasserstein distance is minimized by backpropagating through the Sinkhorn algorithm. SAE models the aggregated posterior as an implicit distribution and therefore does not need a reparameterization trick for gradients estimation. Moreover, it requires virtually no adaptation to different prior distributions. We demonstrate its flexibility by considering models with hyperspherical and Dirichlet priors, as well as a simple case of probabilistic programming. SAE matches or outperforms other autoencoding models in visual quality and FID scores. …

Bordered Blocked Diagonal Form (BBDF)
This paper presents a distributed simulation based method for harmonic resonance assessment (HRA) in multi-area large-scale power systems. Further consideration is devoted to the early harmonic frequency-scan formulation to shape them into a Bordered Blocked Diagonal Form (BBDF), which is suitable for parallel processing. The proposed algorithm (BBDF) allows operator of each area of an interconnected system to independently conduct the HRA. A large-change sensitivity based approach is then handled in a secure platform to apply the effects of whole network to each single area. The introduced decentralized HRA is capable to find the exact values as those of the interconnected system through TCP/IP communication media. The developed method is successfully implemented in an existing software package and applied to IEEE 14-bus harmonic test system, followed by a discussion on results. …

Deep Feature Selection Using Paired-Input Nonlinear Knockoffs (DeepPINK)
Deep learning has become increasingly popular in both supervised and unsupervised machine learning thanks to its outstanding empirical performance. However, because of their intrinsic complexity, most deep learning methods are largely treated as black box tools with little interpretability. Even though recent attempts have been made to facilitate the interpretability of deep neural networks (DNNs), existing methods are susceptible to noise and lack of robustness. Therefore, scientists are justifiably cautious about the reproducibility of the discoveries, which is often related to the interpretability of the underlying statistical models. In this paper, we describe a method to increase the interpretability and reproducibility of DNNs by incorporating the idea of feature selection with controlled error rate. By designing a new DNN architecture and integrating it with the recently proposed knockoffs framework, we perform feature selection with a controlled error rate, while maintaining high power. This new method, DeepPINK (Deep feature selection using Paired-Input Nonlinear Knockoffs), is applied to both simulated and real data sets to demonstrate its empirical utility. …

Gradient Normalization
Deep multitask networks, in which one neural network produces multiple predictive outputs, are more scalable and often better regularized than their single-task counterparts. Such advantages can potentially lead to gains in both speed and performance, but multitask networks are also difficult to train without finding the right balance between tasks. We present a novel gradient normalization (GradNorm) technique which automatically balances the multitask loss function by directly tuning the gradients to equalize task training rates. We show that for various network architectures, for both regression and classification tasks, and on both synthetic and real datasets, GradNorm improves accuracy and reduces overfitting over single networks, static baselines, and other adaptive multitask loss balancing techniques. GradNorm also matches or surpasses the performance of exhaustive grid search methods, despite only involving a single asymmetry hyperparameter $\alpha$. Thus, what was once a tedious search process which incurred exponentially more compute for each task added can now be accomplished within a few training runs, irrespective of the number of tasks. Ultimately, we hope to demonstrate that direct gradient manipulation affords us great control over the training dynamics of multitask networks and may be one of the keys to unlocking the potential of multitask learning. …

HyperKron Graph
Graph models have long been used in lieu of real data which can be expensive and hard to come by. A common class of models constructs a matrix of probabilities, and samples an adjacency matrix by flipping a weighted coin for each entry. Examples include the Erd\H{o}s-R\'{e}nyi model, Chung-Lu model, and the Kronecker model. Here we present the HyperKron Graph model: an extension of the Kronecker Model, but with a distribution over hyperedges. We prove that we can efficiently generate graphs from this model in order proportional to the number of edges times a small log-factor, and find that in practice the runtime is linear with respect to the number of edges. We illustrate a number of useful features of the HyperKron model including non-trivial clustering and highly skewed degree distributions. Finally, we fit the HyperKron model to real-world networks, and demonstrate the model’s flexibility with a complex application of the HyperKron model to networks with coherent feed-forward loops. …

Logical Positivism
Logical positivism and logical empiricism, which together formed neopositivism, was a movement in Western philosophy whose central thesis was verificationism, a theory of knowledge which asserted that only statements verifiable through empirical observation are cognitively meaningful. …

Kubeflow
The Machine Learning Toolkit for Kubernetes. The Kubeflow project is dedicated to making deployments of machine learning (ML) workflows on Kubernetes simple, portable and scalable. Our goal is not to recreate other services, but to provide a straightforward way to deploy best-of-breed open-source systems for ML to diverse infrastructures. Anywhere you are running Kubernetes, you should be able to run Kubeflow. …

Cross Adversarial Source Separation (CASS)
This paper introduces a cross adversarial source separation (CASS) framework via autoencoder, a new model that aims at separating an input signal consisting of a mixture of multiple components into individual components defined via adversarial learning and autoencoder fitting. CASS unifies popular generative networks like auto-encoders (AEs) and generative adversarial networks (GANs) in a single framework. The basic building block that filters the input signal and reconstructs the $i$-th target component is a pair of deep neural networks $\mathcal{EN}_i$ and $\mathcal{DE}_i$ as an encoder for dimension reduction and a decoder for component reconstruction, respectively. The decoder $\mathcal{DE}_i$ as a generator is enhanced by a discriminator network $\mathcal{D}_i$ that favors signal structures of the $i$-th component in the $i$-th given dataset as guidance through adversarial learning. In contrast with existing practices in AEs which trains each Auto-Encoder independently, or in GANs that share the same generator, we introduce cross adversarial training that emphasizes adversarial relation between any arbitrary network pairs $(\mathcal{DE}_i,\mathcal{D}_j)$, achieving state-of-the-art performance especially when target components share similar data structures. …

Diagrammatic Reasoning
Diagrammatic reasoning is reasoning by means of visual representations. The study of diagrammatic reasoning is about the understanding of concepts and ideas, visualized with the use of diagrams and imagery instead of by linguistic or algebraic means.
Can We Automate Diagrammatic Reasoning? …

Agnostic Federated Learning
A key learning scenario in large-scale applications is that of federated learning, where a centralized model is trained based on data originating from a large number of clients. We argue that, with the existing training and inference, federated models can be biased towards different clients. Instead, we propose a new framework of agnostic federated learning, where the centralized model is optimized for any target distribution formed by a mixture of the client distributions. We further show that this framework naturally yields a notion of fairness. We present data-dependent Rademacher complexity guarantees for learning with this objective, which guide the definition of an algorithm for agnostic federated learning. We also give a fast stochastic optimization algorithm for solving the corresponding optimization problem, for which we prove convergence bounds, assuming a convex loss function and hypothesis set. We further empirically demonstrate the benefits of our approach in several datasets. Beyond federated learning, our framework and algorithm can be of interest to other learning scenarios such as cloud computing, domain adaptation, drifting, and other contexts where the training and test distributions do not coincide. …

Modular Attention Network (MAttNet)
In this paper, we address referring expression comprehension: localizing an image region described by a natural language expression. While most recent work treats expressions as a single unit, we propose to decompose them into three modular components related to subject appearance, location, and relationship to other objects. This allows us to flexibly adapt to expressions containing different types of information in an end-to-end framework. In our model, which we call the Modular Attention Network (MAttNet), two types of attention are utilized: language-based attention that learns the module weights as well as the word/phrase attention that each module should focus on; and visual attention that allows the subject and relationship modules to focus on relevant image components. Module weights combine scores from all three modules dynamically to output an overall score. Experiments show that MAttNet outperforms previous state-of-art methods by a large margin on both bounding-box-level and pixel-level comprehension tasks. …

Neural Attentive Knowledge-Based Model (NAK)
Grade prediction for future courses not yet taken by students is important as it can help them and their advisers during the process of course selection as well as for designing personalized degree plans and modifying them based on their performance. One of the successful approaches for accurately predicting a student’s grades in future courses is Cumulative Knowledge-based Regression Models (CKRM). CKRM learns shallow linear models that predict a student’s grades as the similarity between his/her knowledge state and the target course. A student’s knowledge state is built by linearly accumulating the learned provided knowledge components of the courses he/she has taken in the past, weighted by his/her grades in them. However, not all the prior courses contribute equally to the target course. In this paper, we propose a novel Neural Attentive Knowledge-based model (NAK) that learns the importance of each historical course in predicting the grade of a target course. Compared to CKRM and other competing approaches, our experiments on a large real-world dataset consisting of $\sim$1.5 grades show the effectiveness of the proposed NAK model in accurately predicting the students’ grades. Moreover, the attention weights learned by the model can be helpful in better designing their degree plans. …

Averaged Gradient Episodic Memory (A-GEM)
In lifelong learning, the learner is presented with a sequence of tasks, incrementally building a data-driven prior which may be leveraged to speed up learning of a new task. In this work, we investigate the efficiency of current lifelong approaches, in terms of sample complexity, computational and memory cost. Towards this end, we first introduce a new and a more realistic evaluation protocol, whereby learners observe each example only once and hyper-parameter selection is done on a small and disjoint set of tasks, which is not used for the actual learning experience and evaluation. Second, we introduce a new metric measuring how quickly a learner acquires a new skill. Third, we propose an improved version of GEM (Lopez-Paz & Ranzato, 2017), dubbed Averaged GEM (A-GEM), which enjoys the same or even better performance as GEM, while being almost as computationally and memory efficient as EWC (Kirkpatrick et al., 2016) and other regularization-based methods. Finally, we show that all algorithms including A-GEM can learn even more quickly if they are provided with task descriptors specifying the classification tasks under consideration. Our experiments on several standard lifelong learning benchmarks demonstrate that A-GEM has the best trade-off between accuracy and efficiency.
➘ “Gradient Episodic Memory” …

Domain Style Transfer Network (DSTN)
Collecting a large-scale and well-annotated dataset for image processing has become a common practice in computer vision. However, in the ancient painting area, this task is not practical as the number of paintings is limited and their style is greatly diverse. We, therefore, propose a novel solution for the problems that come with ancient painting processing. This is to use domain transfer to convert ancient paintings to photo-realistic natural images. By doing so, the ancient painting processing problems become natural image processing problems and models trained on natural images can be directly applied to the transferred paintings. Specifically, we focus on Chinese ancient flower, bird and landscape paintings in this work. A novel Domain Style Transfer Network (DSTN) is proposed to transfer ancient paintings to natural images which employ a compound loss to ensure that the transferred paintings still maintain the color composition and content of the input paintings. The experiment results show that the transferred paintings generated by the DSTN have a better performance in both the human perceptual test and other image processing tasks than other state-of-art methods, indicating the authenticity of the transferred paintings and the superiority of the proposed method. …

Interval-based Prediction Uncertainty Bounding (IPUB)
The problem of machine learning with missing values is common in many areas. A simple approach is to first construct a dataset without missing values simply by discarding instances with missing entries or by imputing a fixed value for each missing entry, and then train a prediction model with the new dataset. A drawback of this naive approach is that the uncertainty in the missing entries is not properly incorporated in the prediction. In order to evaluate prediction uncertainty, the multiple imputation (MI) approach has been studied, but the performance of MI is sensitive to the choice of the probabilistic model of the true values in the missing entries, and the computational cost of MI is high because multiple models must be trained. In this paper, we propose an alternative approach called the Interval-based Prediction Uncertainty Bounding (IPUB) method. The IPUB method represents the uncertainties due to missing entries as intervals, and efficiently computes the lower and upper bounds of the prediction results when all possible training sets constructed by imputing arbitrary values in the intervals are considered. The IPUB method can be applied to a wide class of convex learning algorithms including penalized least-squares regression, support vector machine (SVM), and logistic regression. We demonstrate the advantages of the IPUB method by comparing it with an existing method in numerical experiment with benchmark datasets. …

ModelarDB
To monitor critical infrastructure, high quality sensors sampled at a high frequency are increasingly installed. However, due to the big amounts of data produced, only simple aggregates are stored. This removes outliers and hides fluctuations that could indicate problems. As a solution we propose compressing time series with dimensions using a model-based method we name Multi-model Group Compression (MMGC). MMGC adaptively compresses groups of correlated time series with dimensions using an extensible set of models within a user-defined error bound (possibly zero). To partition time series into groups, we propose a set of primitives for efficiently describing correlation for data sets of varying sizes. We also propose efficient query processing algorithms for executing multi-dimensional aggregate queries on models instead of data points. Last, we provide an open-source implementation of our methods as extensions to the model-based Time Series Management System (TSMS) ModelarDB. ModelarDB interfaces with the stock versions of Apache Spark and Apache Cassandra and thus can reuse existing infrastructure. Through an evaluation we show that, compared to widely used systems, our extended ModelarDB provides up to 11 times faster ingestion due to high compression, 65 times better compression due to the adaptivity of MMGC, 92 times faster aggregate queries as they are executed on models, and close to linear scalability while also being extensible and supporting online query processing. …

Trained Rank Pruning
The performance of Deep Neural Networks (DNNs) keeps elevating in recent years with increasing network depth and width. To enable DNNs on edge devices like mobile phones, researchers proposed several network compression methods including pruning, quantization and factorization. Among the factorization-based approaches, low-rank approximation has been widely adopted because of its solid theoretical rationale and efficient implementations. Several previous works attempted to directly approximate a pre-trained model by low-rank decomposition; however, small approximation errors in parameters can ripple a large prediction loss. As a result, performance usually drops significantly and a sophisticated fine-tuning is required to recover accuracy. We argue that it is not optimal to separate low-rank approximation from training. Unlike previous works, this paper integrates low rank approximation and regularization into the training. We propose Trained Rank Pruning (TRP), which iterates low rank approximation and training. TRP maintains the capacity of original network while imposes low-rank constraints during training. A stochastic sub-gradient descent optimized nuclear regularization is utilized to further encourage low rank in TRP. The TRP trained network has low-rank structure in nature, and can be approximated with negligible performance loss, eliminating fine-tuning after low rank approximation. The methods are comprehensively evaluated on CIFAR-10 and ImageNet, outperforming previous compression methods using low rank approximation. …

Risk-Constrained RL Framework
The classic objective in a reinforcement learning (RL) problem is to find a policy that minimizes, in expectation, a long-run objective such as the infinite-horizon discounted or long-run average cost. In many practical applications, optimizing the expected value alone is not sufficient, and it may be necessary to include a risk measure in the optimization process, either as the objective or as a constraint. Various risk measures have been proposed in the literature, e.g., mean-variance tradeoff, exponential utility, the percentile performance, value at risk, conditional value at risk, prospect theory and its later enhancement, cumulative prospect theory. In this article, we focus on the combination of risk criteria and reinforcement learning in a constrained optimization framework, i.e., a setting where the goal to find a policy that optimizes the usual objective of infinite-horizon discounted/average cost, while ensuring that an explicit risk constraint is satisfied. We introduce the risk-constrained RL framework, cover popular risk measures based on variance, conditional value-at-risk and cumulative prospect theory, and present a template for a risk-sensitive RL algorithm. We survey some of our recent work on this topic, covering problems encompassing discounted cost, average cost, and stochastic shortest path settings, together with the aforementioned risk measures in a constrained framework. This non-exhaustive survey is aimed at giving a flavor of the challenges involved in solving a risk-sensitive RL problem, and outlining some potential future research directions. …

spamGAN
Online reviews have become a vital source of information in purchasing a service (product). Opinion spammers manipulate reviews, affecting the overall perception of the service. A key challenge in detecting opinion spam is obtaining ground truth. Though there exists a large set of reviews online, only a few of them have been labeled spam or non-spam. In this paper, we propose spamGAN, a generative adversarial network which relies on limited set of labeled data as well as unlabeled data for opinion spam detection. spamGAN improves the state-of-the-art GAN based techniques for text classification. Experiments on TripAdvisor dataset show that spamGAN outperforms existing spam detection techniques when limited labeled data is used. Apart from detecting spam reviews, spamGAN can also generate reviews with reasonable perplexity. …

Distribution Regression
Linear regression is a fundamental and popular statistical method. There are various kinds of linear regression, such as mean regression and quantile regression. In this paper, we propose a new one called distribution regression, which allows broad-spectrum of the error distribution in the linear regression. Our method uses nonparametric technique to estimate regression parameters. Our studies indicate that our method provides a better alternative than mean regression and quantile regression under many settings, particularly for asymmetrical heavy-tailed distribution or multimodal distribution of the error term. Under some regular conditions, our estimator is $\sqrt n$-consistent and possesses the asymptotically normal distribution. The proof of the asymptotic normality of our estimator is very challenging because our nonparametric likelihood function cannot be transformed into sum of independent and identically distributed random variables. Furthermore, penalized likelihood estimator is proposed and enjoys the so-called oracle property with diverging number of parameters. Numerical studies also demonstrate the effectiveness and the flexibility of the proposed method. …

Sequential Association Rule Mining That can Extract Hierarchical Structure of Tasks in Reinforcement Learning (SARM-HSTRL)
Reinforcement learning (RL) techniques, while often powerful, can suffer from slow learning speeds, particularly in high dimensional spaces. Decomposition of tasks into a hierarchical structure holds the potential to significantly speed up learning, generalization, and transfer learning. However, the current task decomposition techniques often rely on high-level knowledge provided by an expert (e.g. using dynamic Bayesian networks) to extract a hierarchical task structure; which is not necessarily available in autonomous systems. In this paper, we propose a novel method based on Sequential Association Rule Mining that can extract Hierarchical Structure of Tasks in Reinforcement Learning (SARM-HSTRL) in an autonomous manner for both Markov decision processes (MDPs) and factored MDPs. The proposed method leverages association rule mining to discover the causal and temporal relationships among states in different trajectories, and extracts a task hierarchy that captures these relationships among sub-goals as termination conditions of different sub-tasks. We prove that the extracted hierarchical policy offers a hierarchically optimal policy in MDPs and factored MDPs. It should be noted that SARM-HSTRL extracts this hierarchical optimal policy without having dynamic Bayesian networks in scenarios with a single task trajectory and also with multiple tasks’ trajectories. Furthermore, it has been theoretically and empirically shown that the extracted hierarchical task structure is consistent with trajectories and provides the most efficient, reliable, and compact structure under appropriate assumptions. The numerical results compare the performance of the proposed SARM-HSTRL method with conventional HRL algorithms in terms of the accuracy in detecting the sub-goals, the validity of the extracted hierarchies, and the speed of learning in several testbeds. …

Binary Matching Pursuit
We study the problem of learning latent feature models (LFMs) for tensor data commonly observed in science and engineering such as hyperspectral imagery. However, the problem is challenging not only due to the non-convex formulation, the combinatorial nature of the constraints in LFMs, but also the high-order correlations in the data. In this work, we formulate a tensor latent feature learning problem by representing the data as a mixture of high-order latent features and binary codes, which are memory efficient and easy to interpret. To make the learning tractable, we propose a novel optimization procedure, Binary matching pursuit (BMP), that iteratively searches for binary bases via a MAXCUT-like boolean quadratic solver. Such a procedure is guaranteed to achieve an? suboptimal solution in O($1/\epsilon$) greedy steps, resulting in a trade-off between accuracy and sparsity. When evaluated on both synthetic and real datasets, our experiments show superior performance over baseline methods. …

IFSAD
Ensemble learning for anomaly detection of data structured into complex network has been barely studied due to the inconsistent performance of complex network characteristics and lack of inherent objective function. In this paper, we propose the IFSAD, a new two-phase ensemble method for anomaly detection based on intuitionistic fuzzy set, and applies it to the abnormal behavior detection problem in temporal complex networks. First, it constructs the intuitionistic fuzzy set of single network characteristic which quantifies the degree of membership, non-membership and hesitation of each of network characteristic to the defined linguistic variables so that makes the unuseful or noise characteristics become part of the detection. To build an objective intuitionistic fuzzy relationship, we propose an Gaussian distribution-based membership function which gives a variable hesitation degree. Then, for the fuzzification of multiple network characteristics, the intuitionistic fuzzy weighted geometric operator is adopted to fuse multiple IFSs and to avoid the inconsistent of multiple characteristics. Finally, the score function and precision function are used to sort the fused IFS. Finally we carried out extensive experiments on several complex network datasets for anomaly detection, and the results demonstrate the superiority of our method to state-of-the-art approaches, validating the effectiveness of our method. …

Quantum Low Entropy based Associative Reasoning (QLEAR learning)
In this paper, we propose the classification method based on a learning paradigm we are going to call Quantum Low Entropy based Associative Reasoning or QLEAR learning. The approach is based on the idea that classification can be understood as supervised clustering, where a quantum entropy in the context of the quantum probabilistic model, will be used as a ‘capturer’ (measure, or external index), of the ‘natural structure’ of the data. By using quantum entropy we do not make any assumption about linear separability of the data that are going to be classified. The basic idea is to find close neighbors to a query sample and then use relative change in the quantum entropy as a measure of similarity of the newly arrived sample with the representatives of interest. In other words, method is based on calculation of quantum entropy of the referent system and its relative change with the addition of the newly arrived sample. Referent system consists of vectors that represent individual classes and that are the most similar, in Euclidean distance sense, to the vector that is analyzed. Here, we analyze the classification problem in the context of measuring similarities to prototype examples of categories. While nearest neighbor classifiers are natural in this setting, they suffer from the problem of high variance (in bias-variance decomposition) in the case of limited sampling. Alternatively, one could use machine learning techniques (like support vector machines) but they involve time-consuming optimization. Here we propose a hybrid of nearest neighbor and machine learning technique which deals naturally with the multi-class setting, has reasonable computational complexity both in training and at run time, and yields excellent results in practice. …

Hyperbolic Recommender System
Many well-established recommender systems are based on representation learning in Euclidean space. In these models, matching functions such as the Euclidean distance or inner product are typically used for computing similarity scores between user and item embeddings. This paper investigates the notion of learning user and item representations in Hyperbolic space. In this paper, we argue that Hyperbolic space is more suitable for learning user-item embeddings in the recommendation domain. Unlike Euclidean spaces, Hyperbolic spaces are intrinsically equipped to handle hierarchical structure, encouraged by its property of exponentially increasing distances away from origin. We propose HyperBPR (Hyperbolic Bayesian Personalized Ranking), a conceptually simple but highly effective model for the task at hand. Our proposed HyperBPR not only outperforms their Euclidean counterparts, but also achieves state-of-the-art performance on multiple benchmark datasets, demonstrating the effectiveness of personalized recommendation in Hyperbolic space. …